Investigating the factor structure of the Montreal Cognitive Assessment: a qualitative review

The Montreal Cognitive Assessment (MoCA) is one of the most widely used screening instruments for Mild Cognitive Impairment (MCI) and dementia. Despite its popularity, uncertainty remains regarding its factorial structure and psychometric functioning across populations and cultures. This review aims to critically evaluate the factorial validity and dimensionality of the MoCA through Classical Test Theory (CTT) and Item Response Theory (IRT) models.

Following the PICO framework, a qualitative review was conducted using PubMed, Web of Science, PsycINFO, and Google Scholar. Inclusion criteria consisted of peer-reviewed empirical studies employing exploratory or confirmatory factor analyses, as well as IRT in samples of older adults.

Across CTT studies, findings ranged from two-factor to hierarchical multi-factor models, with a general cognitive factor frequently emerging. IRT analyses generally supported a unidimensional latent structure, identifying Executive Function, Visuospatial, and Language items as the most discriminative, while Orientation and Memory showed low discriminative power.

Our results showed that the MoCA primarily measures a general cognitive dimension, reflecting variable contributions from different cognitive domains. Standardizing scoring metrics and ensuring cross-cultural factorial equivalence are essential to enhance the tool's accuracy and interpretation of its score.